There are many issues similar to mine, but solutions to those do not work for me. My speculation is that due having previously installed cuda-5.5 something is still set to that version, but no idea what exactly.
Below: Update after symbolic linking.
Basically, when I run in Python
>>> from pycuda.driver import autoinit Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/home/dawid/anaconda2/lib/python2.7/site-packages/pycuda/driver.py", line 5, in <module> from pycuda._driver import * # noqa ImportError: libcurand.so.5.5: cannot open shared object file: No such file or directory My setting:
.bashrc: export PATH=/usr/local/cuda-8.0/bin:$PATH export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-8.0/lib64 Confirmation:
dawid@junkyard:~$ ldconfig -p | grep libcurand libcurand.so.8.0 (libc6,x86-64) => /usr/local/cuda-8.0/lib64/libcurand.so.8.0 libcurand.so (libc6,x86-64) => /usr/local/cuda-8.0/lib64/libcurand.so dawid@junkyard:~$ ll /usr/local/cuda* lrwxrwxrwx 1 root root 19 Mar 15 10:13 /usr/local/cuda -> /usr/local/cuda-8.0/ /usr/local/cuda-8.0: {bins, docs, nvcc, etc.} NVCC version:
dawid@junkyard:~$ nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2016 NVIDIA Corporation Built on Tue_Jan_10_13:22:03_CST_2017 Cuda compilation tools, release 8.0, V8.0.61 After performing symbolic linking there seem to be additional trait in the trail. Now Python complains about undefined symbol to which Google suggests taking care of Boost library. Any thoughts?
>>> import pycuda.autoinit Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/home/dawid/anaconda2/lib/python2.7/site-packages/pycuda/autoinit.py", line 2, in <module> import pycuda.driver as cuda File "/home/dawid/anaconda2/lib/python2.7/site-packages/pycuda/driver.py", line 5, in <module> from pycuda._driver import * # noqa ImportError: /home/dawid/anaconda2/lib/python2.7/site-packages/pycuda/_driver.so: undefined symbol: _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED1Ev 
